Sepp Hochreiter’s Quest to Kick OpenAI from Language Modelling Supermarket

“As a successor of LSTM. We have a new thing. It’s not published, it’s hidden. It’s called XLSTM,” says the German computer scientist Sepp Hochreiter
Researchers Combine Brain-Like Neurons and FPTT For Faster Neural Nets

The approach can be used to train networks with over 6 million neurons
No Surprise, Andrej Karpathy Returns to OpenAI

With the back-and-forth praising and acknowledgement of each other’s work since ChatGPT’s launch, Karpathy’s jump to OpenAI was long due.
Meet Phraser, the AI Tool that Prompts Neural Networks

What makes this smart search feature exciting is the effortlessness in allowing users to search directly through prompts, eliminating the fuss of keywords
Using ‘Cocktail Party Problem’ to Talk with Animals

Understanding the language of animals and communicating with them is one of the longest-running fields of study in technology and biological sciences alike.
What is special in DeepMind’s Sonnet library for constructing neural nets?

Sonnet creates high-level networks that are easier to train and test with multiple applications.
How NeuTigers deploys deep learning in edge AI for easy covid detection

CovidDeep harnesses the grow-and-prune DNN synthesis paradigm to improve accuracy.
DeepMind open-sources DM21, an AI model for quantum chemistry

DM21 uses a neural network to approximate the energy function component of Density Functional Theory.
OpenAI’s neural theorem prover can solve Math Olympiad problems

The theorem prover achieved 41.2% vs 29.3% on the miniF2F benchmark, a challenging collection of high-school olympiad problems.
How to Use Learning Rate Annealing with Neural Networks?

Learning rate is an important parameter in neural networks for that we often spend much time tuning it and we even don’t get the optimum result even trying for some different rates.
Team Of Software Engineers At Facebook Releases “Neural Network Compiler” For PyTorch 1.10

A team of software engineers at Facebook, led by Software Engineer Bertrand Maher, recently released a JIT compiler for CPUs based on LLVM, called NNC, for “Neural Network Compiler.”
Is Depth In Neural Networks Always Preferable? This Research Says the Contrary

Non-deep networks could be utilised to create low-latency recognition systems, rather than deep networks.